TWI794416B - Metrology of multi-layer stacks and interferometer system - Google Patents

Abstract

Techniques for removing interferometry signal phase variations caused by distortion and other effects in a multi-layer stack include: providing an electronic processor sample interferometry data acquired for the stack using a low coherence imaging interferometry system; transforming, by the electronic processor, the sample interferometry data to a frequency domain; identifying a non-linear phase variation from the sample interferometry data in the frequency domain, in which the non-linear phase variation is a result of dispersion introduced into a measurement beam by the test sample; and removing the non-linear phase variation from the sample interferometry data thereby producing compensated interferometry data.

Description

Metrology method and interferometer system for multi-layer stacked structure

The present invention relates to a measurement method for a multilayer stacked structure.

Devices for wearable virtual and/or augmented reality (VR/AR) typically employ stacks comprising multiple parallel plates. The surfaces of the parallel plates within the stack can be characterized and coated as waveguides so that when the device is placed in front of the user's eyes, light information from around the device is transmitted and redirected to the eye to produce a superimposition of data or images without hindering normal vision. In some cases, the stack employs many slabs arranged in parallel, where each slab directs a different color of light (eg, red, green, and blue). To maintain high-quality images, it is important to maintain good parallelism between the plates, to ensure that certain surfaces have the desired flatness, and to maintain specific separation distances between plates during manufacturing, among other factors. However, measurement of these parameters can be difficult due to the influence of undesired light reflected back from surfaces or features within the stack.

The present invention relates to a method of metering multilayer stacks.

In general, in some embodiments, the subject matter of the present invention may be practiced by a method comprising: using a low-coherence image interferometer system, providing sample interference data of a sample under test to an electronic processor, Wherein the sample to be tested includes a plurality of layers in a stack; using the electronic processor, converting the sample interference data to a frequency domain; identifying the sample interference data in the frequency domain a nonlinear phase change in which the identified nonlinear phase change is the result of dispersion of a measurement beam incident upon the sample to be measured; The nonlinear phase change is identified, thereby generating a compensated interferometric data.

Implementation of the method may include one or more of the following technical features and/or technical features of other embodiments. For example, in some embodiments, the method includes: guiding the measuring beam along a measuring beam path, incident to the sample to be measured; guiding a reference beam along a reference beam path, incident on a reference surface , wherein the measurement beam and the reference beam come from light generated by a common light source, the light includes a plurality of wavelengths, wherein the sample to be measured is at least partially transparent to these wavelengths; when the reference beam and the measurement beam are respectively incident After reaching the reference surface and the sample to be measured, combining the reference beam and the measurement beam to form an output beam; guiding the output beam to a detector array, the detector array includes a plurality of detector elements; and A plurality of interference signals from the detector array are recorded, each of the interference signals corresponds to a different position on the sample to be tested, and the sample interference data includes the interference signals.

In some embodiments, the step of identifying the nonlinear phase variation of the sample interference data further includes: obtaining an average phase variation from at least a subset of the interference signals in the frequency domain; A fitting function of the mean phase variation, wherein the step of removing the identified nonlinear phase variation comprises: removing the fitting function from the sample interference data in the frequency domain. The fitting function that fits the mean phase change has a quadratic form. The fitting function for fitting the average phase change is a polynomial whose power is greater than twice.

In some embodiments, the method includes: converting the compensated interferometric data back to a time domain, wherein the compensated interferometric data in the time domain comprises complex compensated interferometric signals; and using the electronic processor, processing the The compensated interference data in the time domain is used to determine information related to the sample to be tested. The step of processing the compensated interference data in the time domain to determine information related to the sample further includes: determining a relationship between a first interface and a second interface in the sample a distance between. The step of determining the distance between the first interface and the second interface in the test sample further includes: for each of the compensated interference signals, identifying the first interface corresponding to the test sample a first intensity peak of the interface and a second intensity peak corresponding to the second interface in the sample to be tested; and for each of the complex compensated interference signals, the identified first intensity is obtained The interval between where the peak occurs and where the identified second intensity peak occurs. The method further includes: determining a degree of parallelism between the first interface and the second interface based on the separation obtained for each of the compensated interference signals. The step of processing the compensated interference data in the time domain to determine information related to the test sample further includes: determining a flatness of a first interface in the test sample. The step of processing the compensated interference data in the time domain to determine information related to the test sample further includes: determining a thickness of a first plate in the test sample. The step of processing the compensated interference data in the time domain to determine information about the test sample further includes determining a thickness of a film in the test sample. There are two plates separated by a gap in the sample to be tested, and the step of processing the compensated interference data in the time domain to determine information related to the sample to be tested further includes: determining the two plates The thickness of the gap between the plates. The method further includes generating a three-dimensional topography map of the gap. The method further includes determining an average thickness of the gap.

In some embodiments, the method includes: performing an initial scan of the stack to identify information related to at least one candidate interface location in the stack; information, repositioning an interferometer objective and/or the sample to be measured so that a first interface of the stack is positioned near a focal plane of the measurement beam; and translating the interferometer objective and/or the sample to be measured , extracting the sample interference data such that the first interface passes through the focal plane. The step of performing the initial scan includes: relatively translating the interferometer objective lens and/or the test sample; during the relative translation of the interferometer objective lens and/or the test sample, recording the Each of these interference signals corresponds to a different position on the sample to be measured, and is measured by an interference fringe frequency (interference fringe frequency) sub-Nyquist frequency (sub-Nyquist frequency) sampling; and determining the at least one candidate interface position according to the interference signals. The step of performing the initial scan includes: positioning the sample to be measured in a first position relative to the interferometer objective; performing a first translation of the interferometer objective and/or the sample to be measured relative to each other; During the first translation, record a plurality of first interference signals from the detector array; position the sample to be tested in a second position relative to the interferometer objective; execute the interferometer objective and/or the detecting a second translation of the samples relative to each other; during the second translation, recording a plurality of second interference signals from the detector array; and based on the first interference signals and the second interference signals, Determine the at least one candidate interface position.

In some embodiments, at least one of the layers in the stack is a glass plate.

In some embodiments, the stack includes a first plate, and a dielectric film formed on a first surface of the first plate.

In some embodiments, the stack includes a first flat plate, and a first diffraction grating formed on a first surface of the first flat plate. The first diffraction grating is an optical coupler for coupling light into the first plate, out of the first plate, or into and out of the first plate. The stack includes a second diffraction grating formed on a second surface of the first plate. The second diffraction grating is an optical coupler for coupling light into the first plate, out of the first plate, or into and out of the first plate.

In some embodiments, the common light source is a white light source.

In some embodiments, a range of wavenumbers over which the nonlinear phase change of the sample interferometric data is identified is obtained using the low-coherence image interferometer system.

In general, in some embodiments, the subject matter of the present invention may be implemented by a system comprising: a low-coherence light source for emitting light having a complex number of wavelengths; an interferometer objective lens for receiving light from the Light from a low-coherence light source, directing a portion of the light as a reference beam along a reference The reference beam path to a reference surface, guide another part of the light as a measurement beam along a measurement beam path to a sample to be measured, and when the reference beam and the measurement beam are respectively from the reference surface and the sample to be measured After measuring the reflection of the sample, the reference beam and the measurement beam are combined into an output beam; a detector array is used to receive the output beam from the interferometer objective lens and generate a sample interference data, the sample interference data Including information related to the sample to be tested, the sample interference data includes a plurality of interference signals, each of which corresponds to a different position on the sample to be tested; an electronic processor, the electronic processor and the detection The detector array communicates with each other, the electronic processor is used to convert the sample interference data into a frequency domain, wherein the electronic processor is further used to identify a nonlinear phase change of the sample interference data in the frequency domain, wherein the nonlinear The phase change is the result of dispersion after the measurement beam is incident on the sample to be measured, and the nonlinear phase change is removed from the sample interference data, thereby generating a compensated interference data.

Implementations of the system can include one or more of the following described features. For example, in some embodiments, the electronic processor is further configured to: obtain an average phase change from at least a subset of the interferometric signals in the frequency domain; and use a fitting function to fit The average phase change. The fitting function has a quadratic form. The fitting function is a polynomial whose power is greater than twice. The electronic processor is further configured to: convert the compensated interferometric data back to a time domain, wherein the compensated interferometric data in the time domain includes complex compensated interferometric signals; and process the compensated interferometric signals in the time domain interference data to determine information related to the sample to be tested. The information related to the sample to be tested includes a distance between a first interface and a second interface in the sample to be tested. Processing the compensated interference data in the time domain to determine the distance between the first interface and the second interface in the sample under test includes: for each of the compensated interference signals, identifying a first intensity peak corresponding to the first interface in the sample to be tested and a second intensity peak corresponding to the second interface in the sample to be tested; and for each of the complex compensated interference signals Alternatively, an interval between the identified position where the first intensity peak appears and the identified position where the second intensity peak appears is obtained. with the test The information about the sample includes the flatness of a first interface in the sample to be tested. The information related to the sample to be tested includes the thickness of a first flat plate in the sample to be tested. There are two flat plates separated by a gap in the sample to be tested, and the information related to the sample to be tested includes the thickness of the gap between the two flat plates. The information related to the sample to be tested includes the thickness of a film in the sample to be tested.

In some embodiments, the interferometer objective comprises a Michelson interferometer objective.

In some embodiments, the interferometer objective comprises a Mirau interferometer objective, a Linnik interferometer objective, or a wide field objective.

In some embodiments, the low-coherence light source includes a white light source.

98, 99, 152: stripes

150: Interference signal

151: Interference pattern

154:Envelope

190: Interference signal

191: object

192: Substrate

193: film

194: interface

195: interface

196: The first interference pattern

197: Second interference pattern

201: light source

202, 203: lens

204: Aperture

205: Light source module

206: Irradiate light

208: Optical splitter

210: Mirau interference objective lens assembly

211: objective lens

212: Reference plate

213: Optical splitter

215: Reference Mirror

220: sample to be tested

222: Reference beam

224: Measuring beam

230: imaging lens

240: Detector

245: Pupil plane

260:Piezoelectric transducer

270:Piezoelectric Actuator

300: components

302, 304, 306, 308: flat panel

310: Gap

312: sample holder

314: surface

316: Measuring beam

318: optical axis

500, 600: process

502, 504, 506, 508, 510, 512: steps

602, 604, 606, 608, 610, 612: steps

1200: Boundary

Figure 1 is an example of a scanning white light interferometry (SWLI) signal.

Figure 2 is an example of a scanning white light interferometer signal for a sample to be measured including a thin film.

Figure 3 is a schematic diagram of a scanning Mirau interferometer.

Figure 4 is a schematic diagram of an optical device.

FIG. 5 is a flow chart of measuring the surface topography of an optical device with a multilayer stack.

Fig. 6 is a flowchart for compensating for dispersion.

FIG. 7A is a schematic diagram of the interference signal on the surface of a glass plate in the time domain.

Fig. 7B is a schematic diagram of the interference signal on the surface of a glass plate in the time domain.

Figure 8 shows the phase variation of an interfering signal around its spectral peak.

Fig. 9A and Fig. 9B respectively show the interference signal of the rear surface of the glass plate before and after dispersion compensation.

Figure 10 is the interference signal obtained by performing a fast scan.

FIG. 11A is a schematic diagram of the interference signal obtained in the detector in the time domain before dispersion compensation.

FIG. 11B is a schematic diagram of the time-domain interference signal in FIG. 11A after dispersion compensation.

Figures 12A and 12B are surface topography images of the first surface and the second surface, respectively, in a stack of glass plates.

Figure 13 is a three-dimensional representation of the thickness of the gap between the first and second surfaces of Figures 12A and 12B.

Referring to FIG. 1, a simulated low coherence interference signal 150 includes complex detector intensity values obtained from a single point on an object (such as a silicon wafer with a single reflective interface) of. The intensity values are plotted as a function of the optical path length difference (OPD), which is the difference between the light reflected from a point on the object and the light reflected from a reference object in the interferometer. The difference between the paths taken. The interference signal 150 is a signal generated by a low coherence scanning interferometry (CSI) scanning optical path difference path, such as moving an optical element and/or object to change the path or reference of light reflected from the object. The path taken by the beam.

Referring to Figure 1, intensity values are plotted as a function of optical path difference (here scan position) and an interference pattern 151 is plotted with a complex number of fringes 152 bounded by a low coherent envelope on either side of the maximum ( low coherence envelope)154 constraints. Without a low coherence envelope, the fringes of the interference pattern will usually show similar amplitudes over a wide range of optical path differences. The low coherence envelope itself does not appear in the interfering signal and is shown for illustrative purposes only. The position of the interference pattern on the optical path difference axis is usually related to the position of zero optical path difference (zero OPD), such as reflection from a point on the object The scanning position or spatial position of zero optical path difference between the light of the light and the light reflected from the reference object. The scanning position of zero optical path difference is a function of the surface topography of the object. The surface topography of the object is used to describe the relative height of each point on the object, as well as the orientation and position of the object itself. The scanning position of zero optical path difference affects The position of each point on the object relative to the interferometer. In some embodiments, the interference signal also includes components contributed by the sample to be measured, such as dispersion and absorption caused by intervening layers between materials.

The width of the low coherence envelope 154 of the modulated fringe 152 amplitude is generally related to the coherence length of the received detection light. The factors that determine the coherence length are the temporal coherence phenomenon (temporal coherence phenomenon) and the spatial coherence phenomenon (spectral coherence phenomenon). For example, the temporal coherence phenomenon is related to the bandwidth of the light source, and the spatial coherence phenomenon is related to the light irradiating the object. Depending on the angle of incidence. Generally speaking, as (1) increasing the bandwidth of the light source, and/or (2) increasing the range of incident angles, the coherence length decreases accordingly. Depending on the configuration of the interferometer used to acquire the data, the overall performance of coherence phenomena may be dominated by one of them, or several of them simultaneously contribute substantially to the overall coherence length. The coherence length of the interferometer can be determined by obtaining the interference signal from an object with a single reflective surface (eg, not a thin-film structure). The interference length corresponds to modulating the full width at half maximum of the envelope of the observed interference pattern.

As shown in FIG. 1, the interference signal 150 results from the detection of light having a range of optical path differences that vary by a magnitude greater than the width of the coherence envelope and, therefore, greater than the coherence length of the detected light. In general, low coherence interference signals can be generated by obtaining interference fringes amplitude-modulated by the coherence envelope of the detected light. For example, when the amplitudes of the observed interference fringes differ by at least 20%, 30%, or 50%, an interference pattern can be obtained along the path of the optical path difference. For example, the peak amplitude of fringe 98 is about 50% smaller than the peak amplitude of fringe 99 .

The low coherence interferometer can detect the interference signal within the optical path difference range, and the optical path difference of the interference signal is greater than or equal to the coherence length of the interferometer. For example, the optical path difference of the detected interference signal may be greater than at least twice the coherence length (e.g., about 3 times or more, 5 times or more) the coherence length above, 10 times or more, 50 times or more, 100 times or more). In some embodiments, the coherence length of the detected light is on the order of the height variation of object features, eg, on the order of a few microns or less, but greater than the nominal wavelength of the detected light.

Referring to FIG. 2 , an interference signal 190 is extracted from an object 191 comprising a substrate 192 and a cover layer such as a thin film 193 . An interface is defined between the substrate 192 and the film. The outer surface 195 of the film defines an interface between the object and its surrounding environment, such as air, other glass, or a vacuum. A general definition of an interface is the change in refractive index between two locations on an object. An object can contain many membranes between other layers.

The interference signal 190 includes a first interference pattern 196 generated by the interface 194 and a second interference pattern 197 generated by the interface 195 . The first interference pattern 196 and the second interference pattern 197 overlap. For example, the maxima of the interference patterns (196, 197) are separated by an optical path difference less than the coherence length of the interferometer, and the interference patterns (196, 197) are not separated by regions of zero intensity. Overlapping interference patterns may produce erroneous results because they distort each other.

For example, the interferometer can be a low-coherence scanning interferometer, including Michelson, Linnik, and Mirau interferometers. Figure 3 shows a scanning interferometer of the Michelson type. The light source module 205 is provided to a beam splitter 208 which directs the illumination light 206 to a Mirau interference objective lens assembly 210 . Illumination light 206 may include broadband light (eg, white light) from a broadband light source having spectral characteristics that produce the desired coherence length. The Mirau interference objective assembly 210 includes an objective lens 211 , a reference plate 212 with a reflective coating on its central portion (which defines a reference mirror 215 ), and a beam splitter 213 . When in operation, the objective lens 211 focuses illuminating light through the reference plate 212 onto the sample 220 to be measured. The beam splitter 213 reflects the first part of the focused light to the reference mirror 215 to define the reference beam 222 , and transmits the second part of the focused light to the sample 220 to define the measurement beam 224 . Then, the beam splitter 213 recombines the measuring beam and the reference beam, the measuring beam is reflected (or scattered) back from the sample 220 to be measured, the reference beam is reflected back from the reference mirror 215, and the objective lens 211 and the imaging lens 230 are used to detect device 240 (for example, In a multi-pixel camera), the combined light interferes with each other to form an image. The detector's measurement signals are sent to a computer (not shown).

The scanning of the embodiment in Fig. 3 involves a piezoelectric transducer (piezoelectric transducer, PZT) 260, and the piezoelectric transducer 260 is coupled to the Mirau interference objective lens assembly 210, allowing the Mirau interference objective lens assembly 210 to be treated as a whole along the optical axis of the objective lens 211. The test sample 220 is scanned, and a scanning interference signal (such as I (ξ, h ) is generated at each pixel of the camera, where I represents the signal intensity of the interference data, and ξ is the interferometer scanning orthogonal to the object surface coordinates, and h is the height of the surface). In other embodiments, piezoelectric transducers are coupled to the sample under test, rather than to the Mirau interference objective assembly 210, to provide relative movement between the sample under test and the Mirau interference objective assembly 210, as shown. Piezoelectric actuator (PZT actuator) 270 . In some embodiments, one or both of the reference mirror 215 and the beam splitter 213 can be moved along the optical axis of the objective lens 211 .

The light source module 205 includes a spatially extended source (spatially extended source) 201, a telescope formed by lenses 202 and 203, and an aperture 204 located in the front focal plane of the lens 202 (the position of the aperture 204 and the rear focal plane of the lens 203 coincide). The above configuration images a spatially extended source onto the pupil plane 245 of the Mirau interference objective assembly 210, which is an example of Koehler imaging. The size of the aperture 204 controls the size of the illuminated field on the sample 220 to be tested. In some embodiments, the light source module may include a configuration in which a spatially extended source is directly imaged onto the sample to be tested, which is referred to as critical imaging. Any type of light source module can be used with other types of interferometers, such as Linnik type scanning interferometers.

FIG. 4 is a schematic diagram of an example of a device 300 that may be used in an artificial reality (AR)/virtual reality (VR) device. Element 300, also referred to herein as a test sample, comprises a plurality of layers arranged in a stack. In this example, the layers include plates 302 , 304 , 306 , 308 . The plate is held in place by the sample holder 312 . One or more panels within the stack are spaced from adjacent panels in the stack by respective gaps (eg, gap 310 ). Tablets can also be stacked Diffractive optical elements and/or coatings (not shown) formed on either end of the stack, or on individual plates throughout the stack are included. Diffractive optical elements can be used as optical couplers (e.g., holographic input and output couplers) for coupling light into and/or out of individual plates, which in turn can serve as a means for moving light waveguide.

In some embodiments, for example, for AR/VR applications, it is useful to identify and control the parallelism of the plates within element 300, as well as to measure other properties of the device, such as the surface flatness of one or more plates, one or more The surface roughness of the multiple plates, and the distance between the plates (eg, gap thickness). Coherent Scanning Interferometer (CSI) as a measurement technique for measuring stacked devices (e.g. element 300) offers several advantages, for example, since CSI typically uses continuous broadband light sources (e.g. light emitting diodes, halogen lamps , incandescent light source, etc.), the incident light spectrum can be tailored directly using standard optical components such as filters. Furthermore, in CSI, the broadband spectrum naturally suppresses cavity interference whose optical path difference exceeds the interference length of the light source, which is of particular concern in optical devices with layers separated by gaps.

Nonetheless, using CSI to measure the surface topography of an optical element having multiple layers separated by gaps, such as element 300, can be challenging for several reasons. For example, when interfaces are brought close together, the proximity of the interfaces may produce overlapping interference patterns, resulting in distortion of the interference signal. In some cases, multiple intermediate layers increase the amount of interference signal dispersion and absorption. For example, with relatively large stacks of glass plates, light may be lost due to absorption and scattering, resulting in attenuation of peak amplitudes within the interference signal. In some cases, dispersion may be a result of stresses caused by surface features and/or coatings and the stack assembly process. For example, in some embodiments glass sheets are coated with holographic input and output couplers and glued to the assembly, which can cause the sheets to bend and possibly even touch each other. The dispersion caused by this effect manifests itself as a phase delay in the measured signal. If the dispersion is non-linear, the net effect may be loss of edge contrast and broadening of the interference envelope in low interference signals. The thicker the stack of layers, the more severe the dispersion effect can be as the interferometer scans deeper into the stack. Since the aforementioned nonlinear properties of the signal depend on the dispersion from all intermediate layers in the stack as well as from the dispersion characteristics of the surface features Performing any kind of pre-calibration to compensate for non-linearities is challenging since this information may not be known until the measurement is performed.

The present invention discloses techniques and systems for performing measurements on such multilayer stacks while compensating for non-linearities that cause dispersion and/or other detrimental effects on interfering signals. In general, in some embodiments, techniques encompassed by the present invention include: 1) converting sample interference data from a test sample having a multilayer stack to a frequency domain; 2) identifying the sample interference data at A nonlinear phase change in the frequency domain, wherein the identified nonlinear phase change is the result of dispersion of a measuring beam incident on the sample to be measured; and 3) from the sample interference data, remove the The nonlinear phase change is identified in the frequency domain, thereby generating a compensated interferometric data. This compensated interferometric data is then converted back to the time domain, from which information about the sample under test, such as surface topography and surface separation, can be obtained. In addition, measurements can be taken from compensated interferometric data before conversion back to the time domain.

FIG. 5 is a flowchart illustrating a process 500 for measuring the surface topography of an optical element (eg, element 300 ) having a multilayer stack using a CSI interferometer, such as the Mirau interferometer of FIG. 1 . As previously mentioned, the light source 201 of the Mirau interferometer may be a broadband light source, including light sources with a relatively large frequency bandwidth (e.g., about 100 nm or greater), and may include, for example, light emitting diodes, halogen lamps, arc lamps, incandescent lights etc. For example, the light source may be in the visible portion of the electromagnetic spectrum (eg, white light).

In step 502 , the measuring beam from the light source 201 is guided along the measuring beam path, and is incident on the sample to be tested 220 , the sample to be tested 220 includes, for example, an element 300 with multiple layers arranged in a stacked manner. For example, as shown in FIG. 4 , the measurement beam 316 is directed toward the first surface 314 of the component 300 along the optical axis 318 . Optical axis 318 may be normal to first surface 314 of element 300 . As the measurement beam 316 passes through the element 300, a portion of the measurement beam 316 is reflected at each interface between regions of different refractive indices.

As previously mentioned, the layers of element 300 may include at least Translucent slab. For example, the flat panel may comprise a glass panel. One or more layers within the stack are spaced from adjacent layers in the stack by corresponding gaps, as shown in FIG. 4 . The gap thickness (e.g., the average distance between the opposing faces of two adjacent plates in the stack) can be between about several hundred nanometers (nm) to about several millimeters (mm) (e.g., at least greater than about 100 nm, at least greater than about 1 micrometer (μm), at least greater than about 10 μm, at least greater than about 100 μm, at least greater than about 1 mm). The thickness of one or more flat plates in the stack can range from hundreds of microns to tens of millimeters or more (e.g., at least greater than about 100 μm, at least greater than about 500 μm, at least greater than about 1 mm, at least greater than about 5 mm, at least greater than about 10 mm ). In some embodiments, the plates are coated with a single or multiple layers of one or more different materials. For example, the flat panel may include one or more antireflective or highly reflective coatings consisting of a single layer or multiple layers of dielectric materials with alternating refractive indices. In some embodiments, the surface of the plate contains specific features. For example, grating (eg diffraction grating) etching can be used to make the surface of the panel contain holographic optical couplers. In some embodiments, the first surface of the panel includes a first optical coupler and the second pair of side surfaces of the panel includes a second optical coupler. Optical couplers can be used to couple light into, out of, or into and out of the plate. Each of the plurality of plates may include a membrane and/or a coupler as described herein.

Referring again to FIG. 3, the light from the broadband light source 201 is split by the beam splitter 213 into 1) a first portion which is focused to a reference mirror 215 to define a reference beam 222 in the reference beam path; and 2) a second portion , the second part is focused on the sample to be measured 220 to define a measuring beam 224 in the measuring beam path. In step 504, the reference beam 222 is directed along a reference beam path, incident on a reference surface (eg, the reference plate 212). In step 506, the measurement beam reflected (or scattered) from the sample to be measured 220 is combined with the reference beam from the reference mirror 215 to form an output beam. The imaging lens 230 and the objective lens 211 are used to make the combined light interfere with each other and form an image in the detector 240 (eg, a multi-pixel camera). In some embodiments, a beam splitter may be used such that the reference and measurement beams pass through an equal amount of glass, thereby reducing dispersion associated with contrast broadening. The measurement signal from the detector is sent to a computer (not shown). Interference occurs when the optical path difference between the two optical paths falls within the coherent length of the illuminating light. In order for each surface of the sample to be tested 220 to interfere, a motorized stage is used to scan the sample to be tested 220 and/or components of the interferometer along the optical axis 318 (see FIG. 4 ) such that the sample to be tested 220 One or more surface samples 220 to be tested meet the coherence condition again.

In step 508, a plurality of interference signals from the detector array are recorded, wherein each of the interference signals corresponds to a different position on the sample to be tested and is recorded by a different detector unit, the sample interference data includes the interference The signal is used to provide sample interference data. In the measurement of this embodiment, the sample to be tested 220 is fixed in an adjustable stand that aligns the nominal surface normal of the sample to be tested 220 with the optical axis 318 of the interferometer. The alignment process may include, for example, imaging light source reflections from the sample under test and light source reflections from the reference surface onto the detector elements of detector 240 . Once aligned, the stage is set such that the stack surface closest to the interferometer is just outside (and to the right of) the object space focal plane. The stage then moves the part towards the interferometer at a constant speed while the camera acquires the interferometric pattern. As each surface passes through the equal path area, interference occurs and the camera records the interference pattern. The interferometer can be configured such that the best focus falls at the object plane satisfying the equal optical path condition between the reference and measurement beams.

At step 510, the detector captures interferometric images of different stage positions, and the interferometric images are then provided to one or more electronic processors (not shown), which in turn, at step 512, perform Compensate for dispersion and process the corrected interference pattern using a CSI method such as peak coherence contrast detection, least squares (template) analysis, or frequency domain analysis. Further information on CSI methods for surface topography analysis can be found, for example, in US Pat. For example, the interference pattern from each surface can be analyzed to obtain the surface topography, assuming the scan speed, camera rate and average wavelength of the illumination light are known. If the interference pattern is captured continuously throughout the scan, the relationship information between all surfaces can be preserved.

FIG. 6 is a flowchart showing a process 600 executable by one or more processors to compensate for dispersion and other effects caused by a sample under test that affect the shape of the interference pattern. In a first step 602, one or more processors of the interferometer system transform the sample interferometric data from the time domain to the frequency domain.

For example, without wishing to be limited by theory, the discretely sampled low-coherence interference signal I can be expressed as: the non-coherent sum of interference patterns within a certain frequency range K , as follows:

The frequency range is the sum of the bandwidth of the light source of the interferometer system and the geometric effect of the non-coherent illumination light at the nonzero numerical aperture (nonzero numerical aperture (NA)). In formula (1), ξ is the scanning coordinate of the interferometer orthogonal to the surface of the object, N is the number of samples captured during scanning, q is the Fourier coefficient, j is the index number of the detection element of the detector array, And z, ν are the index numbers of scanning position ξ and frequency k , respectively. Assuming uniform sampling in the scanning range around the entire envelope of the interference signal I , q _j,y can be expressed as follows:

For example, Fourier coefficients can be obtained by performing a forward Fourier transform on the interferometric signal. The Fourier coefficients representing the interferometric data can be expressed in complex form as wavenumbers and corresponding phases.

In some cases, the DC component from the interference data is removed before converting the interference data to the frequency domain. Additionally, interference signals corresponding to selected surfaces of the sample to be measured are isolated. For example, an interference signal from the interference data having a signal strength value higher than a predetermined value can be identified as corresponding to an interface of a selected surface of the sample to be tested. By setting the values of other remaining interfering signals to zero, interfering signals having greater than (or greater than or equal to) predetermined signal strength values can be isolated. Furthermore, time-domain interferometric signals (e.g., isolated interferometric signals) can be time-shifted before transforming them into the frequency domain, e.g., for each pixel, the signal data in the time domain can be time-shifted ,make The peak amplitude of the obtained signal occurs at the beginning of the data set. This removes the linear phase term in the Fourier components.

After converting the sample interference data to the frequency domain, at step 604, nonlinear phase changes of the sample interference data in the frequency domain are identified. As described herein, the nonlinear phase change is the result of dispersion of the measurement beam after it is incident on a layer in the sample under test. Identification of the nonlinear phase change involves deriving the nonlinear phase change from the sample interferometric data itself, rather than from calibration information or from expected characteristics of the sample under test. Identifying nonlinear phase changes may include, for example, at step 606, obtaining phase changes of at least a subset of the plurality of interfering signals in the frequency domain. For example, in Fourier transformed data, the most useful information may be contained in regions of relatively large Fourier coefficients. Accordingly, the step of obtaining the average phase change (step 606) from at least a subset of the interferometric signals may include using one or more processors to calculate and solve for the phase of a subset of Fourier components of sufficiently large amplitude, e.g., selecting Fourier components having at least a wave number that reaches a predetermined signal-to-noise ratio (S/N ratio). Using the wavenumber range obtained by the low-coherence image interferometer system, the phase variation of the sample interferometric data is identified in the wavenumber range. In some embodiments, the step of obtaining the phase change includes: obtaining an average phase change (eg, average, mode, median) from at least a subset of the interference signals (step 606 ). In some implementations, the phase change can be derived individually and independently for each position in the field. However, if the dispersion is uniform over the field, averaging multiple locations can reduce measurement error. Reference is made to, for example, US Pat. No. 5,398,113 (eg, 9:44-10:54) and US Pat. No. 7,522,288 (eg, 11:49-13:12), each of which is incorporated by reference in its entirety for information on evaluating phase information for interferometric data. into this article.

The step of identifying the nonlinear phase variation further includes: obtaining a fitting function for fitting the phase variation (step 608). For example, as described above, an average phase change for a selected wavenumber can be derived for all pixels or a subset of pixels in a region. Dispersion induced by material interlayers and/or stresses can often lead to quadratic phase nonlinearities in the sample under test. Therefore, the function to fit the phase change can have a quadratic form, eg x2. However, the nonlinear phase change can have forms other than quadratic. In addition, other functions can also be used to fit phase changes, such as functions with polynomials with powers greater than twice, exponential functions, logarithmic functions, spline fitting, Gaussian fitting, etc.

After the nonlinear phase changes are identified, in step 610, the identified nonlinear phase changes of the sample interferometric data are removed in the frequency domain to generate compensated interferometric data. For example, removing the identified nonlinear phase variation (step 610) may include, for each pixel or subset of pixels, subtracting a fitting function (eg, best fit) from the sample interference data. In some embodiments, the identified nonlinear phase information can be further analyzed to provide useful information, including material properties such as group velocity index and/or thickness if the material of the layer is known, and if material defects is known, then analyze the pollutants that affect the absorption and dispersion.

After removing the nonlinear phase variation, at step 612 the compensated sample interferometric data can be converted back to the time domain. One or more processors may then process the compensated interference pattern using CSI methods to provide measurement information such as surface topography, as described herein. In this way, the interference signal itself can be used to evaluate the influence of dispersion on the interference signal in the sample to be measured, and no further information (eg refractive index, dispersion properties or layer thickness) of the interlayer material is required. Furthermore, measurement information including topography maps can be derived from compensated interferometric data in the frequency domain using frequency domain methods (eg, as described in US Pat. No. 5,398,113, the contents of which are hereby incorporated by reference in their entirety). Although this embodiment assumes that the dispersion properties are the same within the field, this procedure can be extended to apply to field dependence (field- dependent) dispersion.

In an embodiment, the processor may determine the distance between the first interface and the second interface in the sample to be tested according to the compensated interference data. Determining the distance between the first interface and the second interface may include, for example, for each compensated interference signal of the plurality of compensated interference signals, identifying a first intensity peak corresponding to the first interface in the sample under test and a corresponding The second intensity peak of the second interface in the sample to be tested. Then, for each compensated interference signal of the plurality of interference signals, the one or more processors may find the location where the identified first intensity peak occurs and the location where the identified second intensity peak occurs. The interval between the current positions. The average distance between the different compensated interference signals can be calculated to provide the average distance between the first interface and the second interface. This distance may correspond to, for example, the average gap thickness between layers within the sample to be tested. For example, the distance may be the average gap thickness between the first and second flat plates within the sample to be tested. Alternatively, the distance may correspond to the average thickness of the layers formed on the flat plate in the sample to be measured. For example, the distance may correspond to the thickness of a thin film dielectric layer formed on the surface of the panel under test. Alternatively, the distance may correspond to the thickness of a flat plate in the sample to be tested.

In some embodiments, one or more processors determine the degree of parallelism between the first interface and the second interface based on the separation obtained for each of the compensated interference signals. For example, one or more processors can be used to output distance data for each pixel into a topography map showing whether the thickness of the gap between two flat plates within the sample under test is uniform or non-uniform. For example, a topography map can show the difference in topography of the faces of the first interface and the second interface. Output the topography map to a display. In some embodiments, the one or more processors are configured to determine other information based on the topography, including, for example, the root-mean-square difference between the first interface and the second interface, the first The peak-valley difference between the interface and the second interface, or any other parameterization of the topography map. In some embodiments, one or more processors can determine the flatness of one or more interfaces within a sample to be tested. For example, the surface form of an interface in a sample to be measured can be obtained relative to a reference plane of the interferometer system.

In some embodiments, the amount of data acquired using the techniques disclosed in the present invention may be very large and time consuming, especially for samples with relatively thick plates with gaps between them. For example, for a sample to be tested that is stacked by eight flat plates with a thickness of 0.5 mm, the gap between the flat plates is 50 μm, and the thickness of the entire stack is 4.35 mm. If for each camera frame, the CSI sampling rate along the scan direction is equal to 1/8 of the wavelength (1/4 of the fringe), and the average light source wavelength is 500nm, scanning the stack and obtaining all surfaces requires more than 70,000 camera frames ( camera frames). Assuming each image is 500×500 pixels and digitized in 8 bits, this corresponds to approximately 17.5GB of data material. Also, for a camera operating at 100Hz, the process can take about 700 seconds (11.7 minutes).

In some embodiments, performing a fast initial scan at a higher translation rate can increase the throughput of data retrieval for identifying information about at least one candidate interface location within the stack. Based on information about the position of at least one candidate interface within the test sample, the interferometer objective lens and/or the test sample can be repositioned to position the at least one candidate interface of the stack near the focal plane of the measurement beam. More detailed information can then be obtained by translating the interferometer objective lens and/or the sample under test, so that at least one candidate interface can pass through the focal plane at a slower rate and/or acquire sample interference data at a higher sampling rate.

For example, in some embodiments, fast scans are performed with Sub-Nyquist acquisitions. The Nyquist frequency is the sampling frequency at which each interference fringe has two sampling points. In sub-Nyquist sampling, the sampling frequency will be less than half the frequency of the fringe signal. For example, the scan speed (of the objective lens and/or the sample to be measured) can be increased by an odd multiple (3,5,7,...) (Sub-Nyquist multiple), and the camera will The inverse of is shuttered (1/3,1/5,1/7,...). This reduces acquisition time and data volume by sub-Nyquist multiples. Shuttering ensures that the camera integrates the same phase range for all sub-Nyquist multiples to minimize contrast blur. In some cases, it is necessary to increase the intensity of the light source to compensate for the shutter. Additionally, measurement noise and environmental sensitivity may increase when operating an interferometer system in a sub-Nyquist data acquisition architecture. For additional information on performing fast scanning, see for example US Patent US 5,398,113 and the journal article " High-speed non-contact profiler based on scanning white light interferometry ", L.Deck and P.de Groot, Appl.Opt.33(31 ), 7334-7338 (1994), each of which is incorporated herein by reference in its entirety.

Based on the data obtained using the quick initial scan, at least one candidate interface location within the stack can be identified. At least one candidate interface can be identified by locating a portion of the interference signal with a local peak. For example, if the sample to be tested consists of multiple plates with gaps between adjacent plates, And for those interested in determining the gap between plates, rather than other relevant information about the plate surface, a high-speed scan (sub-Nyquist scan) can be performed on the entire sample to be tested to identify the plateau positions of all interfaces. The stage position corresponding to the interface occurs where the amplitude of the time-domain interferometric signal reaches a local maximum. Peaks of signal amplitudes approximately spaced from each other by the expected plate gap distance can then be marked as candidate surfaces for the plate. A standard scan may then be performed at a slower rate than the initial scan (e.g., at a rate equal to or greater than the Nyquist frequency) to measure the surface defining the identified gap and record a new interferometric signal, The new interference signal covers the facing surfaces of two different plates in the stack. The disclosed dispersion compensation technique can then be performed on the newly recorded interferometric signal and CSI analysis applied to the corrected data to obtain a more accurate position of the two surfaces. If the scanned surfaces are close enough to each other that interference data can be obtained with a single acquisition scan, then the gap thickness variation can be determined by subtracting the difference between the scanned positions corresponding to the surface positions.

In some embodiments, the initial scan profile is used to quickly identify the location of the candidate surface so that the analysis of the candidate surface can be performed more accurately. For example, after a quick initial scan as described herein, the interferometer and/or the sample under test may be repositioned such that the identified candidate surfaces are located near the focal plane of the interferometer system. From this new position, a standard scan at a slower rate than the initial scan (eg, at a rate equal to or greater than the Nyquist frequency) can be performed and a new interferometry signal recorded. The disclosed dispersion compensation techniques can then be performed on the newly recorded interferometric signals, and CSI analysis applied to the corrected data to obtain and output information about candidate surfaces. The above description describes a quick initial scan as being used for initial identification of candidate surfaces, after which a second scan of candidate surfaces can be performed to obtain more detailed information. However, in some embodiments, a quick scan may provide sufficient detail about the candidate surface without performing a second additional scan.

application

The low coherence interferometric measurement method and system described in the present invention can be used for any of the following surface analysis problems: planar surface topography, texture measurement, surface form measurement, correlation between multiple surfaces measurement (thickness and parallelism), surface defect detection, simple films, multi-layer films, stacked multi-layer objects, gaps between layers, sharp edges and surface features that produce diffraction or complex interference effects, unresolved surface roughness, Unresolved surface features such as sub-wavelength width grooves on otherwise smooth surfaces, dissimilar materials, polarization-dependent properties of surfaces, deflection, vibration, or motion of surfaces or deformable surface features that lead to angle-of-incidence dependence of interference phenomena disturbance. In the case of thin films, the variable parameter of interest may be the film thickness, the refractive index of the film, the refractive index of the substrate, or some combination thereof. Specific applications are discussed next.

AR/VR optics

As explained above in this invention, AR/VR applications can use stacks comprising multiple parallel plates, where the surfaces of the parallel plates within the stack have structural features and coatings that act as waveguides, so that when the element is placed in front of the user's eyes, Light information is transmitted and redirected to the eye to produce a superimposition of data or images without blocking normal vision. In order to maintain a high quality image, it is important to achieve good parallelism between the plates, ensure that certain surfaces have the desired flatness, and maintain a specific separation distance between the plates during manufacturing, among other factors. In some cases, because the slabs used in these optics are relatively thick, dispersion effects can result as the interferometer scanning probe moves deeper into the optical slab.

For example, Figures 7A and 7B show time-domain interference signals obtained using a single detector pixel for a parallel glass plate with a thickness of 6.25 mm. To obtain the data shown in Figures 7A and 7B, the interferometer system is a wide field objective design such as that disclosed in US Pat. No. 8,045,175, the subject matter of which is incorporated herein by reference in its entirety. The interferometer system uses a stepper motor stage to move uniformly at a speed of 500 μm/s. The light source is a 10W dental blue light LED (Dental Blue light LED), its average wavelength is 460nm, and its full width at half maximum is about 25nm. This light source provides an approximately Gaussian-shaped dispersion-free contrast envelope with a standard deviation of 3.5 microns.

Figures 7A and 7B are when a single pixel of the detector in the interferometer system was captured Space-domain interferometric data, corresponding to the front and back surfaces of the optical slab, respectively. In other words, the interferometer scans from the front surface to the back surface of the glass plate. The horizontal axis represents the sample number and corresponds to the scanning position (for each sample signal, its initial position is reset to 0), while the vertical axis represents the signal intensity. Each signal in Figures 7A and 7B represents a data sample obtained with a scanning motion of approximately 57.5 nm and at the Nyquist frequency. It can be clearly seen from Fig. 7B that the coherence width of the interference signal on the back surface is obviously widened. This is due to the dispersion caused by the intervening glass between the front and rear surfaces of the glass sheet. As shown in Fig. 7B, the rear surface signal has a smaller interference peak observed around the sample 500, which is due to the spatial coherence characteristic of the light source, and does not mean that the rear surface reflection is weaker.

Using the techniques disclosed in this invention, the interferometric signal is converted to the frequency domain, where phase variations are identified and removed. For example, Figure 8 shows the phase variation around the spectral peak (corresponding to wavenumber lattice 0) of the back-surface interference signal. In Fig. 8, the horizontal axis corresponds to the spectral wavenumber bin, and the vertical axis corresponds to the phase (unit: radian). From Figure 8, it can be clearly seen that the spectral peak of the rear surface interference signal exhibits a quadratic phase change. The phase change is due to dispersion induced by the intervening glass between the front and rear surfaces of the glass plate. A quadratic function 800 is used to fit the phase variation, and then the quadratic function 800 is subtracted from the frequency domain signal.

Then Fourier transform is performed on the dispersion-corrected frequency-domain signal to obtain the corrected time-domain interference signal. Figures 9A and 9B show interference signals observed from the rear surface of a glass plate before and after dispersion compensation. As shown in Fig. 9B, dispersion compensation has little effect on spatial coherence peaking.

As explained herein, in some cases, a fast initial scan may be performed to quickly identify interfaces within the multilayer stack, eg, corresponding to gap locations between optical slabs. The actual scan is performed on an optical element like element 300 shown in FIG. 4 . Use the same system as in Figure 7A and Figure 7B to obtain the quick scan profile. Although element 300 is shown with a stack of 4 flat plates, in the experiments below, the stack used has at least 6 parallel glass plates each with a thickness of about a few hundred microns. thickness. The gap between adjacent plates is at least 10 microns. Starting from a first side of the stack (eg, surface 314 of element 300), the stack is first translated through the focal plane of the interferometer using a fast scan. Specifically, the fast scan is performed at a rate of about 140 micrometers per second (e.g., about 20 times the typical scan rate for a 100 Hz camera, and about 20 times the typical scan rate for an interferometer operating at visible wavelengths). sampling twice the Nyquist limit) to acquire interferometric data over a small area of the stack to identify locations corresponding to the surfaces of the glass plates within the stack.

Figure 10 shows the interference signal obtained by performing a fast scan. The horizontal axis represents the scanning position (unit: mm), and the vertical axis represents the amplitude of the interference signal. As shown in FIG. 10, the interference signal includes a first local peak at about 0.1 mm, which corresponds to the initial surface of the measurement beam incident stack (similar to surface 314 in component 300). After the first peak, local peaks (eg, at about 0.75 mm, about 1.4 mm, about 2.1 mm, about 2.75 mm, and about 3.4 mm) represent two closely spaced surfaces. In practice, the interference signal at the above locations should appear as two closely spaced peaks, one for each surface of a glass plate. Dispersion effects cause the peaks to blur together (eg, at 2.75mm and 3.4mm) as the scan goes further into the element.

After identifying the surface profile of the plate (as shown in Figure 10), the sample stage is repositioned at each surface to obtain the surface topography for each surface or pair of surfaces so that short CSI scans can be obtained from each interference of a surface or pair of surfaces.

For example, after repositioning the stage so that it is in front of the 8th and 9th surfaces (corresponding to the surfaces of the 4th and 5th plates, respectively) about 2.75mm inside the stack, perform a 150 µm long, 3X SubNyquist CSI scan for interference from two surfaces. The system has an average wavelength of 460nm and a scan increment of 172.5nm between camera frames for 3X SubNyquist CSI scanning. Figure 11A shows the raw time-domain interference signal of two closely spaced 8th and 9th surfaces, before dispersion compensation, as observed by a pixel of the detector in the interferometer system. As shown in Fig. 11A, before dispersion compensation, the interference signals from each interface are merged together, and it is difficult to distinguish the individual interference features from the two surfaces. No. Figure 11B illustrates the same time-domain interference signal after performing dispersion compensation as disclosed in the present invention. As shown in Figure 11B, the peaks corresponding to the different interfaces can now be easily resolved. As shown by the data in Figure 11B, the spacing between the surfaces is about 25 microns.

In some cases, interference signals representing surfaces are identified for each pixel in the field and analyzed using a CSI algorithm. For example, for two surfaces at about 2.7 mm in Figure 10, a peak contrast algorithm is applied to each pixel in the predetermined field to generate a topography map of the surfaces. Figure 12A is a topographical view of the first surface (surface 8 within the stack corresponding to the rear surface of the fourth glass plate). Figure 12B is a topographical view of the second surface (surface 9 within the stack corresponding to the rear surface of the fifth glass plate). Surface 8 actually comprises embossed features, a distinct step 1200 can be observed in Figure 12A, shifting the contrast envelope due to the complex optical properties of the features. Since the profile of the two surfaces is acquired in a single scan, their relative orientation is preserved and the gap between them can be calculated from their difference. As shown in Figure 13, the gap can be represented in three dimensions, where the z-axis corresponds to the gap thickness.

Other surfaces or pairs of surfaces within the stack may also be measured as described herein until all surfaces of interest have been measured. Although the scan disclosed herein is to illuminate the test sample from a first side and then translate the test sample through the focal plane, in some embodiments it may be advantageous to perform a scan from the first side and partially through the test sample/stack of. Then, perform a second partial scan from the second phase side of the test sample/stack, through the test sample/stack. For example, this can be achieved by reversing the orientation of the sample/stack within the sample holder after the first scan and before performing the second scan. Performing a scan in this manner may be useful when the transmission properties of the surface and material are very poor, avoiding performing a single scan on the test sample/stack that would be too noisy near the interface near the test sample/stack end .

digital realization

The technical characteristics of the information processing described in the present invention can be digital electronic circuit, computer hardware body, firmware or a combination of the above. These features can be embodied in a computer program product in an information carrier, for example, in a machine-readable storage device, allowing a programmable processor to execute; these technical features can be executed by a programmable processor executing an instruction program, and the input Materials operate on and produce output to perform the functions of the described implementation. The above-mentioned technical features can be implemented in one or more executable computer programs on a programmable system, which includes at least one programmable processor, which is coupled to receive data and instructions from a data storage system, and Data and instructions are transmitted to at least one input device and at least one output device. A computer program consists of a set of instructions that directly or indirectly perform a certain activity or bring about a certain result in a computer. A computer program may be written in any form of programming language, including compiled or interpreted languages, in a computing environment, and may be arranged in any form, including as a stand-alone program or as a module, component, subroutine or other suitable unit for use.

Suitable processors for the execution of a program of instructions include, by way of example, general and special purpose microprocessors, one of the many processors of any kind of computer. Generally, a processor will receive instructions and data from read only memory or random access memory, or both. A computer includes a processor for executing instructions and one or more memories for storing instructions and qualifications. In general, a computer also includes, or is operatively coupled to communicate with, one or more mass storage devices for storing data files; these mass storage devices include magnetic Disks, such as: internal hard disks and removable disks; magneto-optical disks; and optical disks. Storage devices suitable for embodying computer program instructions and data include all forms of non-volatile memory including: for example semiconductor memory devices such as EPROM, EEPROM and flash memory devices; magnetic disks such as internal hard drives and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks. Processors and memory can be supplemented by or incorporated in ASICs (Special Application Integrated Circuits). These functions can be implemented in a single process or distributed among multiple processors in one or more locations. For example, these functions may use cloud technology for data transmission, storage and/or analysis.

category

It must be noted that, as used herein and in the appended claims, the singular forms "a", "an" and "the" may include plural referents unless the context otherwise requires Specify, for example, when the singular quantifier "single" (single) is explicitly used.

As used herein, the terms "adapted" and "configured" mean that an element, assembly, or other subject matter is designed and/or intended to perform a given function. Accordingly, the use of "adapt" and "configure" does not mean a specific element, component or other subject matter that is simply "capable of" performing a given function.

As used herein, with respect to a list of more than one entity, "at least one" and "one or more" refer to any one or more entities in the entity list, and are not limited to those specifically listed in the entity list At least one of each entity in . For example, "at least one of A and B" (or "at least one of A or B" or "at least one of A and/or B" with the same meaning) may only mean that A and B are used alone, or that A and A combination of B.

As used herein, the conjunction "and/or" between the first entity and the second entity means (1) the first entity; (2) the second entity and (3) between the first entity and the second entity one of them. Multiple entities listed with "and/or" should be construed in the same fashion, ie, "one or more" of the entities so conjoined. Other entities may optionally be present other than the entities specifically identified by the "and/or" clause, whether related or unrelated to those entities specifically identified.

While this specification contains many specific implementation details, these should not be construed as limitations on the scope of any invention or of what may be claimed, but rather as descriptions of features specific to particular embodiments of particular inventions.

Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

Furthermore, although the above features may be described as functioning by certain combinations, and even if initially stated as such, in some cases, one or more features of the claimed combination may be removed from the combination. Certain features are removed, and claimed combinations mean subcombinations or variations of subcombinations.

Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as necessarily requiring that operations be performed in the particular order shown or sequentially, or that all illustrated operations be performed, to achieve desirable results . In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system elements in the above-described embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems may generally be integrated into a single software product or disassembled. Into a variety of packaged software products.

Embodiments of the claimed subject matter of the present invention are as described above. The following claims include other embodiments. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multiplexing and parallel processing may be preferred.

Many embodiments of the invention are described above. However, it should be understood that various modifications may be made without departing from the spirit and scope of the invention. Accordingly, other embodiments are within the scope of the following claims.

600: process

602: Convert sample interference data from time domain to frequency domain

604: In the frequency domain, identify the phase change of the interference data

606: Obtain phase changes of at least a subset of interferometric signals

608: Obtain the fitting function of fitting phase change

610: Remove the identified phase variation from the sample interferometric data to obtain compensated interferometric data

612: Convert the compensated interference data back to the time domain

Claims (38)

A method for metrology of multilayer structures, comprising: using a low coherence image interferometer system, providing sample interference data of a sample to be tested, wherein the sample to be tested includes a plurality of layers in a stack; using the electronic a processor, converting the sample interference data into a frequency domain; identifying a nonlinear phase change of the sample interference data in the frequency domain, wherein the identified nonlinear phase change is a measurement beam incident on the sample to be measured As a result of subsequent dispersion; removing the identified nonlinear phase change in the frequency domain from the sample interference data, thereby producing a compensated interference data; directing the measurement beam along a measurement beam path , incident to the sample to be measured; guiding a reference beam along a reference beam path, incident to a reference surface, wherein the measurement beam and the reference beam come from light generated by a common light source, the light contains complex wavelengths, wherein the The sample to be tested is at least partially transparent to the wavelengths; after the reference beam and the measurement beam are respectively incident on the reference surface and the sample to be tested, the reference beam and the measurement beam are combined to form an output beam; directing the output beam to a detector array comprising a plurality of detector elements; and recording a plurality of interference signals from the detector array, each of the interference signals corresponding to the sample to be tested The sample interference data includes the interference signals at different positions on the stack; an initial scan of the stack is performed to identify information related to at least one candidate interface position in the stack; the information about the position of the interface, repositioning an interferometer objective and/or the sample under test to position a first interface of the stack near a focal plane of the measurement beam; and Translating the interferometer objective lens and/or the sample to be measured to capture interference data of the sample so that the first interface passes through the focal plane. The method described in claim 1, wherein the step of identifying the nonlinear phase variation of the sample interference data further includes: obtaining an average phase from at least a subset of the interference signals in the frequency domain change; and obtaining a fitting function that fits the average phase change, wherein the step of removing the identified nonlinear phase change comprises: removing the fitting function from the sample interference data in the frequency domain Mode. The method described in claim 2, wherein the fitting function for fitting the average phase change has a quadratic form. The method described in item 2 of the scope of the patent application, wherein the fitting function for fitting the average phase change is a polynomial whose power is greater than twice. The method described in claim 1, comprising: converting the compensated interference data back to a time domain, wherein the compensated interference data in the time domain includes complex compensated interference signals; and using the electronic processing A device processes the compensated interference data in the time domain to determine information related to the sample to be tested. The method described in item 5 of the scope of the patent application, wherein the step of processing the compensated interference data in the time domain to determine the information related to the sample to be tested further includes: determining a first one of the samples to be tested A distance between an interface and a second interface. The method described in claim 6, wherein the step of determining the distance between the first interface and the second interface in the sample to be tested further includes: for each of the compensated interference signals , identifying the first interface corresponding to the sample to be tested and a first intensity peak corresponding to the second interface in the sample to be tested; and for each of the complex compensated interference signals, obtaining the identified first intensity peak The interval between the occurrence and the identified occurrence of the second intensity peak. The method described in item 7 of the claimed claims includes determining the parallelism between the first interface and the second interface according to the distance obtained for each of the compensated interference signals. The method described in item 5 of the scope of the patent application, wherein the step of processing the compensated interference data in the time domain to determine the information related to the sample to be tested further includes: determining a first one of the samples to be tested The flatness of an interface. The method described in item 5 of the scope of the patent application, wherein the step of processing the compensated interference data in the time domain to determine the information related to the sample to be tested further includes: determining a first in the sample to be tested The thickness of the plate. The method described in claim 5, wherein the step of processing the compensated interference data in the time domain to determine information related to the sample to be tested further includes: determining a thin film in the sample to be tested the thickness. The method described in claim 5, wherein there are two flat plates separated by a gap in the sample to be tested, and wherein the compensated interference data in the time domain is processed to determine the correlation with the sample to be tested The step of sample-related information further includes: determining the thickness of the gap between the two flat plates. The method described in item 12 of the scope of the patent application further includes generating a three-dimensional topography map of the gap. The method described in item 12 of the scope of the patent application further includes determining one of the gaps The average thickness. The method described in item 1 of the scope of the patent application, wherein the step of performing the initial scanning includes: relatively moving the interferometer objective lens and/or the sample to be measured; During translation, the interference signals from the detector array are recorded, each of the interference signals corresponds to a different position on the sample to be measured and is divided by an interference fringe frequency Sampling at sub-Nyquist frequency; and determining the position of the at least one candidate interface according to the interference signals. According to the method described in item 1 of the patent scope of the application, the step of performing the initial scanning includes: positioning the sample to be measured at a first position relative to the interferometer objective lens; executing the interferometer objective lens and/or the a first translation of the samples relative to each other; during the first translation, a plurality of first interference signals from the detector array are recorded; the sample to be measured is positioned at a second relative to the interferometer objective position; perform a second translation of the interferometer objective lens and/or the sample to be measured relative to each other; during the second translation, record a plurality of second interference signals from the detector array; and according to the The first interference signal and the second interference signals determine the position of the at least one candidate interface. The method of claim 1, wherein at least one of the layers in the stack is a glass plate. The method according to claim 1, wherein the stack includes a first flat plate, and a dielectric film formed on a first surface of the first flat plate. The method as described in claim 1 of the patent claims, wherein the stack includes a first flat plate, and a first diffraction grating formed on a first surface of the first flat plate. The method described in claim 19, wherein the first diffraction grating is an optical coupler for coupling light into the first plate, out of the first plate, or into and into Leave the first plate. The method of claim 19, wherein the stack includes a second diffraction grating formed on a second surface of the first flat plate. The method described in claim 21, wherein the second diffraction grating is an optical coupler for coupling light into the first plate, out of the first plate, or into and into Leave the first plate. The method as described in claim 1 of the patent application, wherein the common light source is a white light source. The method described in item 1 of the scope of the patent application, wherein the low-coherence image interferometer system is used to obtain a wavenumber range, and the nonlinear phase change of the sample interference data is identified within the wavenumber range. An interferometer system comprising: a low-coherence light source for emitting light with complex wavelengths; an interferometer objective lens for receiving light from the low-coherence light source and guiding a part of the light as a reference beam along a reference The beam path is to a reference surface, and another part of the light is guided as a measurement beam along a measurement beam path to a sample to be measured, and when the reference beam and the measurement beam are respectively from the reference surface and the to-be-measured After sample reflection, the reference beam and the measurement beam are combined into an output beam; a detector array is used to receive the output beam from the interferometer objective lens and generate a sample interference related data, the sample interference data includes information related to the sample to be tested, the sample interference data includes a plurality of interference signals, each of the interference signals corresponds to a different position on the sample to be tested; an electronic processor, The electronic processor and the detector array communicate with each other, the electronic processor is used to convert the sample interference data to a frequency domain, wherein the electronic processor is further used to identify the sample interference data in the frequency domain a linear phase change, wherein the nonlinear phase change is the result of dispersion of the measurement beam incident on the sample to be measured, and the nonlinear phase change is removed from the sample interference data, thereby generating a compensated interference data; And wherein the electronic processor is further used to: i) perform an initial scan of a stack to identify information related to the location of at least one candidate interface in the stack; position, repositioning an interferometer objective and/or the sample to be measured to position a first interface of the stack near a focal plane of the measurement beam; and iii) translating the interferometer objective and and/or the sample to be measured, extracting interference data of the sample so that the first interface passes through the focal plane. The system of claim 25, wherein the electronic processor is further configured to: obtain an average phase change from at least a subset of the interference signals in the frequency domain; and use a fitting function The formula fits this average phase change. The system described in claim 26, wherein the fitting function has two secondary form. The system as described in claim 26 of the patent application, wherein the fitting function is a polynomial whose power is greater than twice. The system of claim 25, wherein the electronic processor is further configured to: convert the compensated interferometric data back to a time domain, wherein the compensated interferometric data in the time domain includes complex compensated interferometric data signal; and processing the compensated interference data in the time domain to determine information related to the sample to be tested. The system as described in claim 29, wherein the information related to the sample to be tested includes a distance between a first interface and a second interface in the sample to be tested. The system described in claim 30, wherein processing the compensated interference data in the time domain to determine the distance between the first interface and the second interface in the sample to be tested comprises: For each of the compensated interference signals, identifying a first intensity peak corresponding to the first interface in the test sample and a second intensity peak corresponding to the second interface in the test sample and, for each of the complex compensated interference signals, finding an interval between the identified location where the first intensity peak occurs and the identified location where the second intensity peak occurs. The system as described in claim 29, wherein the information related to the sample to be tested includes the flatness of a first interface in the sample to be tested. The system as described in claim 29, wherein the information related to the sample to be tested includes the thickness of a first flat plate in the sample to be tested. The system of claim 29, wherein in the sample to be tested there are two plates separated by a gap, and wherein the information related to the sample to be tested includes the gap between the two plates thickness of. The system as described in claim 29, wherein the information related to the sample to be tested includes the thickness of a film in the sample to be tested. The system as described in claim 25, wherein the interferometer objective lens includes a Michelson interferometer objective lens. The system as described in claim 25, wherein the interferometer objective lens includes a Mirau interferometer objective lens, a Linnik interferometer objective lens, or a wide field objective lens. The system as described in claim 25, wherein the low-coherence light source includes a white light source.

Images